| QC Report Information | |
|---|---|
| Generated on 19 noviembre 2025 | |
| Metric | Value |
| Report Date | 19 noviembre 2025 |
| Indicator ID | 5650 |
| Indicator Name | Carbon dioxide (CO2) emissions per GDP |
| Data Vintage | 2025-11-19T173458 |
| Data Source | CAIT - WRI |
| Dimension IDs | |
| Dimension Names | |
| Time Coverage | 1990 - 2022 |
| Number of Countries | 35 |
| Total Records | 1149 |
Quality Check Report
Quality Check Report for Indicator 5650
Summary Info
This section summarizes key metadata about the indicator and dataset under review. It helps contextualize the scope of the data quality checks.
Completeness Check
Data Completeness by Country and Year
This section verifies whether the internal dataset covers the same country-year combinations as the public CEPALSTAT version. It flags:
“Missing in Both” (no data in either source)
“Old Only” (data was removed or lost in the new version)
“New Only” (new values added that were not in the original public version)
“Present in Both” (overlap between datasets)
The visual below shows data coverage by country and year. Gaps or additions should be reviewed for justification.
New Values Time Series
This section visualizes the values of the new data points. Whether any of the new data points for the new year are anomalous can be explored in the Country-Level Analysis section.
Country Values Time Series
This shows the same information but is scaled to the country level.
Value Comparisons (Between Data Sets)
This section analyzes changes between values that are available in both the public and internal data sets. Key statistics include mean and absolute differences, and the share of values with changes greater than ±20%. Red dots in the plots identify values that exceed this threshold.
This helps identify outliers, input errors, or significant data revisions that may need explanation.
Summary Statistics
| Change Detection Summary | |
|---|---|
| Key metrics comparing values present in both internal and public datasets | |
| Metric | Value |
| Mean % difference | 26.9% |
| % of values with large differences (<20%) | 56.0% |
| Mean absolute difference | 204.1 |
| Median reported value | 313.9 |
| Minimum reported value | 21.2 |
| Maximum reported value | 1713.6 |
Absolute Value Comparison
This section compares data values for country/year pairs between the public and internal data sets. The dashed line indicates perfect alignment between the two sources. Any points colored in red indicate a greater than 20% difference between the two data sets.
Relative Difference Comparison
This section compares relative differences between the public and internal data. A percent change is calculated between the public data and internal data, and the data is displayed in country/year pairs. The ±20% tolerance threshold band is colored a light green. Large relative changes above 20% are colored red.
Percent Difference Threshold Table
This table displays records where the updated internal values differ from the public CEPALSTAT data by more than 20%, corresponding to the red points in the plots above.
Rows with a z-score (z_perc_diff) greater than 3 are statistically significant outliers, indicating values that deviate strongly from the overall distribution and may require further review or explanation.
Outlier Detection (Within Data Set)
This section evaluates the quality of the internal data independently by providing statistical outlier detection based on z-scores. It flags any records where the percent difference is more than 3 standard deviations from the mean, which could indicate outliers or extreme changes in the data.
A z-score measures how far a value is from the average, in terms of standard deviations. It’s calculated as:
z = (x − μ) / σ
where: x is the value of interest, μ is the mean of the group, and σ is the standard deviation.
Typical interpretations:
A z-score of 0 means the value is at the mean
A z-score of ±1 is within the typical range (68% of data is within 1 s.d.)
A z-score of ±2 is unusual (95% of data is within 2 s.d.)
A z-score of ±3 or more is considered an outlier (99.7% of data is within 3 s.d.)
Country-Level Analysis
The following checks analyze the internal data within each country set. This helps differentiate any significant country-level anomalies that may be less apparent than at the global scale.
Z-Score Threshold Table
This calculates z-scores within each country series to flag major deviations in the individual country series. This displays all entries with a z-score greater than 3. If no entries are shown, then all of the values within each country seem reasonable.
Outlier Country Plots
This section isolates country time series that have raised any flags: either with percent changes exceeding 20% or z-scores exceeding 3. This can be used to visually identify whether these data points are within-country outliers.